A High-Resolution Luminescent Assay for Rapid and Continuous Monitoring of Protein Translocation across Biological Membranes

Protein translocation is a fundamental process in biology. Major gaps in our understanding of this process arise due the poor sensitivity, low time resolution and irreproducibility of translocation assays. To address this, we applied NanoLuc split-luciferase to produce a new strategy for measuring protein transport. The system reduces the timescale of data collection from days to minutes and allows for continuous acquisition with a time resolution in the order of seconds, yielding kinetics parameters suitable for mechanistic elucidation and mathematical fitting. To demonstrate its versatility, we implemented and validated the assay in vitro and in vivo for the bacterial Sec system and the mitochondrial protein import apparatus. Overall, this technology represents a major step forward, providing a powerful new tool for fundamental mechanistic enquiry of protein translocation and for inhibitor (drug) screening, with an intensity and rigor unattainable through classical methods.


Reagents
All chemicals, such as antibiotics, inducers and mitochondrial poisons, were of the highest commercially available grade of purity and were purchased from Sigma-Aldrich. Aqueous solutions were prepared in ultrapure (type I) water (Milli-Q Biocel A10 with pre-treatment via Elix 5, Millipore, Billerica, MA, USA). For non-aqueous solutions, ethanol (99.5%) or dimetylsulfoxide (DMSO), both from Sigma-Aldrich, were used as solvent. To produce yeast expressing 11S in the mitochondrial matrix, the 11S amino acid sequence previously published [3] was codon-optimised for S. cerevisiae and supplemented with the mitochondrial signal sequence of yeast alpha subunit of ATP synthase (ATP1/YBL099W; 1-35aa) on its N-terminus. This gene was purchased on a plasmid from Eurofins (Germany), digested with HindIII and XbaI, and ligated into the corresponding sites of pYES2CT (yielding pYES-mt-11S). The plasmid was verified by sequencing then transformed into YPH499 yeast cells. The mt-11S gene was cloned into a high-copy number plasmid (pYES2) under the control of GAL promoter to facilitate the delivery of high quantities of 11S to the mitochondrial matrix.

His-tagged 11S for proteoliposomes preparations
To produce and purify 11S, the 11S gene (without the mitochondrial signal sequence) was amplified from pYES-mt-11S with a 5' primer containing an NcoI site followed by a 6-his tag (GATCGTCCATGGGCCATCATCATCAT-CATCATGGCGTTTTCACATTGGAG), and a 3' primer containing a HindIII restriction site (GCCTAAAAGCTTC-TAGCTATTGATGGTTACACG). The resulting PCR product was digested with NcoI and HindIII, and ligated into the corresponding sites of pBAD/Myc-His C (yielding pBAD-6H 11S). The plasmid was verified by sequencing then transformed into BL21(DE3) cells.

Tethered 11S for IMV preparations
To produce IMVs with high concentrations of 11S on the inside, we tethered 11S to the periplasmic face of the inner membrane using a lipid anchor [4]. The 11S gene (without the his-tag) was amplified from pBAD-6H 11S using a 5' primer with a NcoI restriction site followed by the signal sequence and first six amino acids of NlpA (EG10657; ACGTAGCCATGGGCAAACTGACAACACATCATCTACGGACAGGGGCCGCATTATTGCTG GCCGGAATTCT-GCTGGCAGGTTGCGACCAGAGTAGCAGCGGCGTTTTCACATTGGAG), and a 3' primer including the HindIII restriction site (CTACGTAAGCTTCTAGCT). The PCR product was then digested with NcoI and HindIII, and ligated into the corresponding sites in pRSFDuet-1. The resulting plasmid was verified by sequencing then co-transformed with pBAD-SecYEG into BL21(DE3) cells.

GST-dark peptide for in-vitro experiments
To allow recombinant expression and purification of large quantities of dark peptide, we fused it to the Cterminus of glutathione-S-transferase (GST). This was done by PCR insertion of a DNA sequence coding for a pep86 version that does not luminesce (dark peptide, see details in [5]) immediately after the BamHI site in pGEX-1 (primer sequences: CATCCTCCAAAATCGGATCCCGGAGTGAGCGGCTG GGCGCTGTTTAAAAAAATTAGC-TAAGAATTCATCGTGACTGAC and GTCAGTCACGATGAATTCTTAGCTAATTTTTTTAAACAGCGCCCAGC-CGCTCACTCCGGGATCCGATTTTGGAGGATG). The resulting plasmid (pGEX-GST-dark) was verified by DNA sequencing, then transformed into BL21 (DE3) cells.

proOmpA(±pep86)
proOmpA with a C-terminal minimal V5 epitope was produced as described previously [6]. For the real-time translocation assays, PCR insertion was used to add the pep86 sequence, preceded by a short GSG linker, after the V5 tag (primers: GAATCCGCTGCTGGGCCTGGGCTCCGGCGTGAGCGGCTG GCGCCTGTTTAAAAAAATTAGC-TAAGCTTACGTAGAACAAAAAC and GTTTTTGTTCTACGTAAGCTTAGCTAATTTTTTTAAACAGGCGCCAGC-CGCTCACGCCGGAGCCCAGGCCCAGCAGCGGATTC). After verifying the clone by DNA sequencing, proOmpA-pep86 was expressed and purified using exactly the same protocol as standard proOmpA.

CytB2±pep86 and ∆mts-CytB2-pep86
An engineered version of yeast cytochrome B2 (YML054C, [7]) comprising the first 158 aa with its hydrophobic domain on the signal sequence deleted (∆43-65) followed by two tandem TEV cleave sites, a Myc tag and a C-term 6xhis tag, was codon-optimised for E. coli and purchased as a gene on plasmid (Eurofins, Germany). Then, the plasmid was digested with NcoI and HindIII, and the insert ligated into the corresponding sites of pBAD/Myc-His C (yielding pBAD-CytB2 ∆43-65 ). For the real-time translocation assays, PCR insertion was used to add the pep86 sequence, preceded by a SGGGGS linker, after the 6xhis tag, yielding pBAD-CytB2 ∆43-65 -pep86. All plasmids were verified by sequencing then transformed into BL21(DE3) cells.

pro11S-GST-dark for in-vivo experiments
To produce bacteria expressing 11S in the periplasm, the amino acid sequence of 11S was codon-optimised for E. coli, supplemented with the signal sequence of bacteria OmpA (EG10669, 1-21 aa) on its N-terminus (pro11S) and purchased as a gene fragment (GeneArt, Invitrogen). The fragment was ligated into pMiniT2.0 according to the manufacturer's instructions (NEB PCR cloning kit, E1203S), and sequence checked with pMiniT F and R primers. pMini-pro11S and pBAD-Myc-His-C were digested with NcoI and HindIII (NEB HF enzymes) according to the manual, fragments agarose gel purified (QIAquick gel extraction kit, #28704), ligated using T4 ligase (ThermoScientific, #EL0011) and transformed into α-select creating pBAD-pro11S.
The GST-dark gene fragment was synthesised (Invitrogen), ligated into pMiniT2.0 according to the manual and sequence checked. pMini-GST-dark and pBAD-pro11S were digested with HindIII and SalI (NEB HF enzymes) according to the manual, fragments agarose gel purified, ligated using T4 ligase and transformed into E. coli α-select creating pBAD-pro11S-GST-dark.

Protein expression and purification
For expression, pre-cultures were inoculated with a single colony of the desired bacterial strain (BL21(DE3) as default) and grown in LB with appropriate antibiotic for 16 h at 37°C, 200 rpm. Cultures were inoculated at 1:100 from pre-cultures in 2xYT plus antibiotic and grown at 37°C, 200 rpm until mid-log phase, then induced for 2.5-3 h with 0.1-0.2% (w/v) arabinose or 1 mM depending on the plasmid. For overexpression of bacterial pre-proteins the MM52 strain was used instead as it contains a temperature-sensitive copy of genomic SecA [8]. When exposed to temperatures above 30°C the mutant SecA is rendered inactive, causing pre-proteins to accumulate in the cytoplasm, typically as inclusion bodies. Therefore, pre-cultures were grown at 30°C and protein expression carried out at 39°C.

His-tagged 11S
Cells were harvested, resuspended in 20 mM Tris pH 7.5, 50 mM KCl (TK) with 10% glycerol (TKG) then cracked open using a cell disruptor (Constant Systems) and clarified by centrifugation. The supernatant was loaded onto a Ni 2+ column packed with chelating Sepharose Fast Flow resin (GE Healthcare), washed in TKG with 50 mM imidazole, then eluted with TKG + 330 mM imidazole. Imidazole was removed by washing with TKG in a spin concentrator, and the final protein concentration determined from A 280 , using the calculated extinction coefficient of 19,940 M −1 .cm −1 . The sample was then snap frozen and stored at -80°C.

GST-dark peptide
Cells were harvested, resuspended in TK buffer then cracked in a cell disruptor and clarified by centrifugation. The supernatant was loaded onto a GSTrap 4B (GE Healthcare) at 4°C and the column washed with TK buffer until A 280 of the flowthrough stopped decreasing. Elution was performed with 10 µM reduced glutathione in TK. The yield of the resulting protein (hereafter GST-dark) was determined from A 280 , using the calculated molar extinction coefficient of 48,360 M −1 .cm −1 . The sample was then snap frozen and stored at -80°C.

pep86-tagged mitochondrial precursors
Cells were harvested, resuspended in TK buffer then cracked in a cell disruptor and clarified by centrifugation. Inclusion bodies were solubilised in TK plus 6 M urea (TK + urea) before loading into an in-house packed Ni 2+ column. After washing with TK + urea, proteins were eluted with 330 mM imidazole in TK + urea and loaded into an in-house packed Q-or S-column. After column wash, proteins were thereafter eluted in TK + urea + 1 M KCl gradient (0-100%) during 20 min. Final fraction was spin concentrated and the final protein concentration determined from A 280 , using the calculated extinction coefficient. The sample was then snap frozen and stored at -80°C.

Pep86-tagged bacterial pre-proteins
For proOmpA, the cell pellet was resuspended in 130 mM NaCl, 20 mM Tris pH 8.0 and cells cracked in a cell disruptor followed by a clarifying spin. A previously established purification protocol was utilised where inclusion bodies were harvested by gentle centrifugation at 4000 g for 15 min and solubilised in urea [6]. The resulting mixture was loaded onto an anion exchange column equilibrated in a salt-free 6 M urea, 10 mM Tris pH 8.0 buffer. A linear salt gradient of 0-1 M was then applied, where proOmpA-pep86 constituted the first protein to elute, with a 280 nm absorbance peak at approximately 40 mM NaCl.
For proSpy, cells were harvested and resuspended in lysis buffer (500 mM NaCl, 50 mM Tris, 30 mM imidazole, pH 8.0) buffer supplemented with cOmplete, EDTA-free Protease Inhibitor Cocktail. The cells were then lysed and clarified by centrifugation. The soluble cell fraction was determined to contain approximately 80% of the total expressed Spy and was therefore loaded onto a 5 mL HiTrap Crude FF nickel affinity chromatography column (GE Healthcare). After washing with lysis buffer, the bound proteins were eluted with lysis buffer containing 300 mM imidazole and then spin-concentrated (5 KDa cut-off) to ∼14 mL before TEV digestion. DTT and EDTA, 1 mM and 0.5 mM respectively, plus ∼0.1 mg/mL TEV protease were added to the suspension and incubated at room temperature for about 3 h. Then, Spy solution was purified by nickel affinity chromatography as described above, but this time collecting the unbound column flow through (His tag removed-Spy). The sample was dialysed overnight into 6 M urea, 20 mM Tris pH 8.0 and then snap frozen and stored at -80°C.

Proteoliposome (PL) preparation
SecYEG proteoliposomes were produced as described previously [6]. Briefly, purified SecYEG was mixed with E. coli polar lipids in DDM, then the detergent removed gradually using BioBeads. To encapsulate 11S, we simply included purified 11S at the desired final concentration (20 µM standard, or as noted in the text) in the SecYEG/polar lipid mix, prior to the addition of biobeads. For the initial experiments, SecYEG/11S PLs were harvested by centrifugation (30 min at 100,000 g) as per the standard method, washed twice by resuspending in 3 mL TKM then centrifuging for 30 min at 100,000 g and pipetting off the supernatant, then resuspended to give the desired final SecYEG concentration. These additional washing steps removed most of the non-encapsulated 11S, reducing the background signal for the transport assays -although they did not obviate the need for GST-dark.
For the 11S concentration series we instead passed the reconstituted PLs over a gravity flow Sephacryl-S1000 column to separate away unbound 11S. These PLs were quantified by scattering at 320 nm (relative to PLs produced using the standard method), then used directly in transport assays. This method is both more effective at removing free 11S and eliminates the centrifugation and resuspension steps, which potentially damage PLs and cause them to leak.

IMV preparation
IMVs were either prepared from E. coli BL21(DE3) or a strain lacking ATP synthase (unc-; HB1 cells [9]), to prevent PMF generation upon addition of ATP. Cells were grown to mid-log phase and 37°C in 2xYT supplemented with 100 µg/mL ampicillin and 50 µg/L kanamycin, then co-induced for 2.5 h with 0.1% arabinose and 1 mM IPTG. Inverted membrane vesicles were prepared from the membranes as described previously [6].

Mitochondrial isolation
mt-11S-expressing yeast were grown overnight at 30°C in synthetic growth media lacking uracil and supplemented with 3% glycerol plus 0.0025% Pen/Strep. Yeast cells were cultured in glycerol-based media to increase mitochondrial mass and maximise mitochondrial function [10] 1% galactose was added at mid-log phase to start inducing mt-11S (total time ∼16h). In the end, mitochondria were isolated through differential centrifugation after cell wall was reduced by 1 mM DTT in 100 mM Tris-SO4 buffer for 15 min at 30°C and then digested with zymolase in sorbitol-phosphate buffer for 30 min at 30°C. The final mitochondrial pellet was resuspended in 250 mM sucrose and 10 mM MOPS, pH 7.2. Mitochondrial protein was quantified by BCA assay, using BSA as standard.

Binding experiments
Complementation of pep86-tagged precursor with pure 6H 11S was performed in 1x Nano-Glo buffer diluted with TK buffer and Prionex (0.1% final). The reaction mix containing 1x furimazine was used to prepare a titration curve of precursor ranging from 13.9 nM to 3 µM which was added to a 96-well plate. Separately, pure 11S was diluted in reaction mix supplemented with 1x furimazine so that automatic injection of 80 µL would give the desired concentration upon addition (30 pM in 100 µL). Luminescence was read on BioTek Synergy Neo2 plate reader (BioTek Instruments, UK) without emission filters every 0.5 s during 30 s at 25°C working on well mode. Obtained data was fitted to a single exponential.

Western blot transport assays
For the Sec-system, western blot transport time courses were performed in a 25°C heat block. Reaction master mixes were prepared in buffer TK + 2 mM MgCl2 (TKM), containing: creatine phosphate to 5 mM, creatine kinase to 0.1 mg/mL, SecYEG IMVs or PLs to 4% of final volume, SecA to 1 µM and proOmpA to 0.2 µM. From each master mix, 5 µL was set aside for a 10% control, and another 50 µl diluted 5-fold into ice-cold 1 mg/mL protease K in 5 mM EDTA as a -ATP (t=0) control. Reactions were immediately started by adding ATP to a final concentration of 1 mM, and 50 µL was taken at various time points and quenched rapidly as for the t = 0 sample. Samples were prepared and western blotted essentially as described previously [6]. Briefly, undigested protein was precipitated on ice for 30 min with 20% (w/v final) trichloroacetic acid, then centrifuged and the supernatant removed. Pellets were dried in a vacuum centrifuge, then resuspended in 30 µL 2.5x NuPage LDS buffer (Thermo Fisher Scientific) and heated to 70°C for 20 min. 10 µL of the resulting samples were run out on a gel, blotted, then developed using an α-v5 primary antibody (SV5-Pk1, GeneTex) and a DyLight 800-labelled secondary antibody (Thermo Fisher Scientific). Western blots were visualised on an Odyssey Fc (LI-COR) and the bands quantified using the built-in software.
For the mitochondrial system, yeast mitochondria (240 µg protein) were diluted in 240 µL import buffer (250 mM sucrose, 80 mM KCl, 5 mM MgCl2 , 10mM K2HPO4 , 10 mM MOPS-KOH pH 7.2) supplemented with 2 mM NADH, 2 mM ATP, 5 mM creatine phosphate and 0.1 mg/mL creatine kinase. The samples were preincubated at 25°C for 5 min before import was started by adding 1 µg/mL substrate. The reaction was allowed to proceed for 1, 3 or 5 min with gentle shaking at 350 rpm before being stopped by the addition of 2.4 µL VOA (containing 100 µM valinomycin, 2 mM Oligomycin and 800 µM antimycin A). Half of each sample was treated with 25 µg/mL Proteinase K on ice for 15 min followed by the addition of 3 mM phenylmethylsulfonyl fluoride (PMSF) for 2-5 min to stop the reaction. Centrifugation at 20,000 g was used to isolate the mitochondria which were then washed with SM buffer (250 mM sucrose, 10 mM MOPS-KOH, pH7.2). Pellets were resolved in 2x sample buffer (4% SDS, 20% glycerol, 125 mM Tris-HCl pH 6.8, 0.02% bromophenol blue, 50 mM DTT) and heated at 65°C for 10 min before being subjected to SDS-PAGE. After blotting, membranes were developed using an anti-myc primary antibody (Cell Signaling) and a DyLight 800-labelled secondary antibody (Thermo Fisher Scientific). Western blots were visualised on an Odyssey Fc (LI-COR) and the bands quantified using the built-in software. Real-time import assays (for the Sec system) were performed at 25°C in a Jobin Yvon Fluorolog (Horiba) with the lamp turned off and emission measured at 460 nm (with slits open to maximum, i.e. 10 nm bandpass). A reaction mix was assembled in a 1 mL cuvette with a stirrer bar by adding (in order): TKM to give a final volume of 1 mL, Prionex (Sigma-Aldrich; registered trademark of Pentapharm AG, Basel) to 0.1%, 10 µL Nano-Glo substrate (furimazine, Promega), creatine phosphate to 5 mM, creatine kinase to 0.1 mg/mL, GST-dark to 40 µM, 1 µL SecYEG/11S HB1(DE3) IMVs or PLs, and SecA to 1 µM. After a 5 min equilibration, a luminescence baseline signal was measured for 1 min, followed by addition of proOmpA-pep86 to 1 µM final concentration. After a further 10 min, ATP was added to 1 mM final concentration, and the transport reaction followed until completion.
The background, caused by association of proOmpA-pep86 with non-encapsulated 11S, fits well to a single (PLs) or double (IMVs) exponential (Fig. 2). Therefore, we fitted the signal from after proOmpA-pep86 but prior to ATP addition to a single exponential, and subtracted the resulting fit from the raw data. This corrected data corresponds to ATP-driven protein transport. When plates were read on a Packard Lumicount BL10001 (Packard BioSciences, Meriden, CT, US), additions were made manually using an 8-channel pipette with manual mixing; if BioTek Synergy Neo2 plate reader (BioTek Instruments, UK) was used instead, additions were made automatically using the system pump set to the default injection speed followed by a 5 s linear shaking step. In both plate readers, luminescence was collected for 0.2 s/well, without emission filters, and the gain was set to allow maximum sensitivity without detector saturation. An initial baseline of 60 sec was acquired before precursor addition and then luminescence was read for at least 20 min. Time between reads was set to the minimum allowed interval in each plate reader -10-12 s for LumiCount and 5 s for Neo2.
1.11. In-vivo β-lactamase secretion assay E. coli MC4100 were transformed with pBAD-pro11S-GST-dark alone or in combination with pSU2718-NDM-1-pep86. Starter cultures were inoculated with a single colony of the desired strain and grown in LB with appropriate antibiotic for 16 h at 30°C, 200 rpm. 5 mL cultures (LB with antibiotic) were inoculated from these starter cultures at 1:100 and grown at 30°C, 200 rpm until OD600∼1. At that point, all cultures were induced by adding arabinose to a final concentration of 0.2% (w/v) and incubated at 39°C. 100 µL of sample was taken at the time of induction and assayed 2 h later using the Nano-Glo Live Cell Assay System (Promega) according to the manufacturer's protocol.
Luminescence was measured over 100 reads during 10-20 min and the data averaged. Background luminescence was deducted from assay data of bla carrying strains.

Data analysis and Statistics
Results are shown as means±SEM of the indicated number of experiments. Apparent rates (k app ) were calculated as the reciprocal of the time it takes to reach half of the maximal luminescent signal (t 50% ). Statistical significance between mean differences was determined using two-tailed Student's t test or one-way ANOVA, when more than two groups were analysed, followed by predefined contrasts using Bonferoni's post-hoc analysis to correct for multiple comparisons. Differences were considered significant if p ¡ 0.05 and categorized accordingly to their interval of confidence. Statistical analyses were performed using Graph Pad Prism version 8.0.0 (GraphPad Software, Inc., San Diego, CA, USA). Abbreviations: IMVs -inner membrane vesicles, POI -protein of interest, WB -western blot. Disclaimer: The list above is not an exhaustive representation of all the methods available at the moment of publication, we would like to acknowledge all the colleagues who have developed assays but are not listed above.  Prof. Agnieszka Chacińska Lab Figure S1: Overview of the new real-time assay to monitor protein translocation on different biological systems. The large 11S fragment was segregated in the mitochondrial matrix, proteoliposome lumen, tethered to the inner membrane of E. coli (inner membrane vesicles) or in the periplasm (in vivo). All pre-protein substrates were tagged on their Cterminus with the high-affinity peptide pep86. To decrease signal background, a non-luminescent high-affinity pep86 peptide (GST-dark) was added to the reactions (mitochondria, E. coli in vitro) or co-expressed in the cytosol for in vivo experiments. In all systems, successful pre-protein translocation was observed as an increase in luminescent signal upon pep86 \ 11S complementation.   Fig. 1c, showing that both SecA and a functional signal sequence are required for import. n=1 for all panels. a b Figure S3: Effects of 11S concentration on pre-protein import traces. Despite the positive correlation between 11S concentration and luminescence signal (left-hand plot) the shape of the curve after normalisation to RLU max remains the same (right-hand plot), confirming that the assay reports protein translocation rather than complementation of both fragments. Figure S4: Effect of GST-dark co-expression with 11S for monitoring NDM1-pep86 translocation in vivo in MC4100 cells. Data is shown as mean ± SEM of three independent experiments. Differences between groups were assessed by Student's t-test.  Figure S5: Detection of mt-11S and quantification in isolated mitochondria. mt-11S mitochondria were isolated by standard differential centrifugation and their protein content was then resolved by SDS-PAGE. 11S was detected using a primary antibody against NanoLuc (gift from Promega, US). a top -increasing amounts of mt-11S can be achieved by controling the amount of galactose during overnight induction; bottom -titration of mt-11S isolated mitochondrial fractions (induced at 1 % galactose) and 6H 11S protein for quantification. b standard curves for 11S chemiluminescence signal as a function of mitochondrial or protein amount. c extrapolation of 11S in mt-11S mitochondria -there is 1.34 µM 11S per mitochondria. d, e relationship between signal of complemented 11S and amount of mitochondria. e is a log 10 -log 2 plot of the linear plot on 'd'. For the experiments in 'd, e' a serial dilution (1:1) of mt-11S mitochondria was prepared in normal reaction buffer in the absence of import (DECA) and GST-dark but in the presence of the detergent NP-40, which releases the content of the mitochondria. The concentration of CytB2 ∆43-65 -pep86 was was chosen to saturate the 11S, 1 µM. The reaction was allowed to equilibrate for 10 min before the addition of furimazine and measurement on the plate reader. Experimental data are shown as mean ± 95% confidence intervals. Dashed area on 'e' represent the limit of detection of the plate reader used (6 million RLU). n=1.  Figure S7: Distinction between import and background traces. Import of CytB2 ∆43-65 -pep86 into NADH-only energised 11S-mitochondria in the absence (green) or presence (coral) of oligomycin, valinomycin and antimycin AA (OVA) cocktail -full pmf dissipation. In the absence of OVA (green) the curve is sigmoidal due to the presence of a lag period during the intial phase of import. Contrarily, the background trace (coral) experiences no lag and resembles a single exponential curve. On the left panel, import was normalised to the maximum RLU of each trace in order to make differences in lag easier to observe. n=1. Figure S8: Panel a and b shows the import of CytB2 ∆43-65 -pep86 into 11S mitochondria in the presence and absence of pmf, respectively. Observationaly, the background signal corresponds to about 1/3 of the RLU max , independently of the concentration of preprotein-pep86 used. Because the signal is observed in the absence of pmf, the driving force for protein import, it reflects extra-mitochondrial events. This is supported by the observation that incubation of GST-dark in the absence of pmf  Figure S9: Pure CytB2 ∆43-65 -pep86 and 6H 11S complementation kinetics (a). To assess the binding affinity of preprotein-pep86 to 11S and its association constant, 6H 11S was kept constant (30 pM) and and the preprotein-pep86 (CytB2 ∆43-65 -pep86) varied from 3 µM to 46 nM. Experiments were setup on a 96 well plate by preparing a serial dilution of CytB2 ∆43-65 -pep86 in NanoGlo buffer supplemented with furimazine following manufacturer's instructions (Promega). Then, complementation was started by injecting a fixed amount of 6H 11S (25 µL) to achieve the desired final concentration in 125 µL. Luminescent signal was measured in a BioTek Synergy Neo2 every 4 sec simultaeously in 8 wells. Average of 2 independent runs are shown in panel b. Obtained data were fitted to a single exponential (black lines in b) and secondary data was plotted in the graph on panel c. Experimental data are shown as mean ± 95% confidence intervals. Error bars are not shown if smaller than the symbol. Fitted data are shown with the 95% confidence band in grey. n=2. Table S5: Sequences of constructs used in this current study.